Hacker News Reader

Comment by amelius | Hacker News Reader

> The short summary is that bad things (no charging) happens if the CC1 and CC2 pins are shorted together anywhere in a USB-C system that is not an audio accessory.

Why should it matter what kind of accessory it is used in?

> Each CC gets its own distinct Rd (5.1 kΩ), and it is important that they are distinct.

Why?

I think this article was not written by an electrical engineer. The reason why this design is inadequate still eludes me. But, RPi should have probably used the reference design, that's clear.

View on HN

> Why should it matter what kind of accessory it is used in?

I suspect you're misunderstanding what they mean by "audio accessory". They don't mean "a standard digital USB device that happens to do audio-related things". USB-C has a special-case allowance to drive analog audio signals through the port (to allow all-analog headphones and such) that has its own CC configuration.

>> Each CC gets its own distinct Rd (5.1 kΩ), and it is important that they are distinct. > Why?

Because of the above; if the CCs use the same Rd, they're shorted together by definition, which isn't allowed to happen. Especially since the USB-C format uses the CC's to determine plug orientation.

>USB-C has a special-case allowance to drive analog audio signals through the port

Wat. Just when I think I've heard it all...

You've seen the Type C - 3.5mm headphone dongles right? They're so tiny and cheap.. did you think they could be active? :)

USB-C tries to be one universal standard that covers everyone's use cases (https://xkcd.com/927/). Some phones used non-standard modes in their USB connectors which output analog audio in the USB pins, to be used with a passive adapter to a 3.5mm socket. So USB-C added, in an optional appendix, a standard mode for that use case. It's completely optional, so you only have to implement it if your device wants to output analog audio in the USB connector, or if your device is one of these passive adapters from USB-C to a 3.5mm socket.

There is also a debug mode that manufacturers can use to send arbitrary debug data connections on the usb-C connections. (manufacturers used to do this over microUSB as well)

> Why?

Because two resistors in parallel (i.e when RD is shorted to RA) lowers the resistance and pulls the voltage down to a range that identifies the device as an audio device to the charger

:edit: A diagram to clear it up https://imgur.com/97OKDf5

Top is as the RPI is effectively configured, Bottom is how it should be.

The author of the article has a doctorate in a relevant field and has worked on USB compliant products, as far as qualifications go he knows what he is talking about.

The link originally pointed to a different article and was updated to mine later hence some slightly confusing comments on here.

On the RPi4 CC1 and CC2 are shorted, while the standard says there should be a resistor (2*5.1kΩ) between them, that's a huge difference.

Reply on news.ycombinator.com